This is interesting: If I do the same test with remote-spawn, it works.
Vista, in a command prompt window:
tsi (node-init (make-node "127.0.0.1" 12346))
in another window:
tsi (node-init (make-node "127.0.0.1" 12345)) (remote-spawn (make-node "127.0.0.1" 12346) (lambda () (raise
'drawbridge))) *** ERROR IN (console)@14.1 -- This object was raised: #<termite-exception #7 origin: #<upid #8 tag: 98C38CA3-14FF-4EE9-84BF-2343401AA10D node: #<node #9 host:"127.0.0.1" port: 12346>> reason: failure object: drawbridge> 1>
I need to play with this some more and see if I can figure out where the exceptions are sent back to the spawning process.
Dave
On Jan 27, David N Murray scribed:
I got the latest termite via the svn command. Gambit-C 4.6.0:
On my BSD machine:
$ tsi
(define (t1)
(spawn (lambda () (raise 'drawbridge))))
(t1)
#<thread #2 anonymous>
[error] Wed Jan 27 14:47:45 EST 2010 #<thread #2 anonymous> *** ERROR IN ##dynamic-env-bind -- This object was raised: drawbridge
On Windows Vista, using the VC version (I created a tsi.bat based on the tsi shell script):
(define (t1)
(spawn (lambda () (raise 'drawbridge))))
(t1)
#<thread #2 anonymous>
(no error or exception is reported)
The same thing happens on Vista with the MinGW Gambit-C build. The exception is not propogated.
All tsi scripts invoke 'gsi -:dar' and the necessary termite include and load commands. I have other termite tests that run fine on the Vista install. It's just that process exceptions are silently eaten.
Exceptions are reported in Vista, just not from termite processes:
(define (t1) (raise 'drawbridge)) (t1)
*** ERROR IN (console)@11.1 -- This object was raised: drawbridge 1>
Thoughts?
TIA, Dave
Gambit-list mailing list Gambit-list@iro.umontreal.ca https://webmail.iro.umontreal.ca/mailman/listinfo/gambit-list